Under present conditions, petroleum refineries are finding it necessary to convert increasingly greater proportions of crude to premium fuels such as gasoline and middle distillates such as diesel and jet fuel. Catalytic cracking processes, exemplified by the fluid catalytic cracking (FCC) process and Thermofor catalytic cracking (TCC) process together, account for a substantial fraction of heavy liquids conversion in modern refineries. Both are thermally severe processes which result in a rejection of carbon to coke and to residual fractions; during catalytic cracking high molecular weight liquids disproportionate into relatively hydrogen-rich light liquids and aromatic, hydrogen-deficient heavier distillates and residues.
Catalytic cracking in the absence of hydrogen does not provide significant desulfurization nor is the nitrogen content of the feed selectively rejected with the coke. Both sulfur and nitrogen therefore concentrate appreciably in the heavier cracking products. Catalytic cracking therefore produces significant quantities of highly aromatic, hydrogen-deficient middle and heavy distillates that have high sulfur and nitrogen levels. Recycling these liquids to the catalytic cracker is often not an attractive option, because they are refractory and difficult to convert and often will impair conversion of the less refractory fresh feed. Generally, the level of heteroatom contaminants increases with the boiling point of the fraction, as shown in Table 1 below which gives the sulfur and nitrogen contents for two FCC product fractions, a light cycle oil and an FCC main column bottoms (proportions and percentages by weight, as in the remainder of this specification unless the contrary is stated).
TABLE 1 ______________________________________ FCC Product Fractions Aromatics pct. S, pct. N, ppm H, pct. ______________________________________ Light Cycle Oil (LCO) 80 3.1 650 9.1 Main Column Bottoms 80 4.6 1500 6.8 ______________________________________
Present market requirements make refractory product streams such as these particularly difficult to dispose of as commercially valuable products. Formerly, the light and heavy cycle oils could be upgraded and sold as light or heavy fuel oil, such as No. 2 fuel oil or No. 6 fuel oil. Upgrading the light cycle oil was conventionally carried out by a relatively low severity, low pressure catalytic hydro-desulfurization (CHD) unit in which the cycle stock would be admixed with virgin mid-distillates from the same crude blend fed to the catalytic cracker.
Currently, however, the refiner is finding a diminished demand for fuel oil. At the same time, the impact of changes in supply and demand for petroleum has resulted in a lowering of the quality of the crudes available to the refiner; this has resulted in the formation of an even greater quantity of refractory cycle stocks. As a result, the refiner is left in the position of producing increased amounts of poor quality cycle streams from the catalytic cracker while having a diminishing market in which to dispose of these streams.
At many petroleum refineries, the light cycle oil (LCO) from the FCC unit is a significant component of the feed to the catalytic hydrodesulfurization (CHD) unit which produces No. 2 fuel oil or diesel fuel. The remaining component is generally virgin kerosene taken directly from the crude distillation unit. The highly aromatic nature of LCO, particularly when the FCC unit is operated in the maximum gasoline mode, increases operational difficulties for the CHD and can result in a product having marginal properties for No. 2 fuel oil or diesel oil, as measured by cetane numbers and sulfur content.
An alternative market for mid-distillate streams is automotive diesel fuel. However, diesel fuel has to meet a minimum cetane number specification of about 45 in order to operate properly in typical automotive diesel engines. Because cetane number correlates closely and inversely with aromatic content, the highly aromatic cycle oils from the cracker typically with aromatic contents of 80% or even higher have cetane numbers as low as 4 or 5. In order to raise the cetane number of these cycle stocks to a satisfactory level by the conventional CHD technology described above, substantial and uneconomic quantities of hydrogen and high pressure processing would be required.
Because of these problems associated with its use as a fuel, recycle of untreated light cycle oil to the FCCU has been proposed as a method for reducing the amount of LCO. Benefits expected from the recycle of LCO include conversion of LCO to gasoline, backout of kerosene from No. 2 fuel oil and diminished use of cetane improvers in diesel fuel. However, in most cases, these advantages are outweighed by disadvantages, which include increased coke make in the FCC unit, diminished quality of the resultant LCO and an increase in heavy cycle oil and gas.
A typical LCO is such a refractory stock and of poor quality relative to a fresh FCC feed that most refineries do not practice recycle of the untreated LCO to any significant extent. One commonly practiced alternative method for upgrading the LCO is to severely hydrotreat prior to recycle to the catalytic cracker or, alternatively, to severely hydrotreat and feed to a high pressure fuels hydrocracker. In both such cases, the object of hydrotreating is to reduce the heteroatom content to low levels while saturating polyaromatics to increase crackability. Although this does enhance the convertibility of these aromatic streams considerably, the economic penalties derived from high hydrogen consumptions and high pressure processing are severe. In addition, in those instances where the production of gasoline is desired, the naphtha may require reforming to recover its aromatic character and meet octane specifications.
Hydrocracking may be used to upgrade the higher-boiling more refractory products derived from catalytic cracking. The catalytic cracker is used to convert the more easily cracked paraffinic gas oils from the distillation unit while the hydrocracker accepts the dealkylated, aromatic cycle oils from the cracker and hydrogenates and converts them to lighter oils. These hydrocracking processes using catalytically cracked feeds either on their own or mixed with virgin feeds have, however, generally been incapable of producing high octane gasoline directly. The reason for this is that they have conventionally been operated at high hydrogen pressures and at relatively high conversion levels so as to maximize the saturation of the aromatics (especially the refractory polynuclear aromatics), removal of heteroatoms in inorganic form and the subsequent conversion of the hydrogenated aromatics to paraffins. While this may produce acceptable diesel fuel (which benefits from the presence of n-paraffins) the octane quality of the gasoline has generally been poor as a consequence of the large quantities of low octane paraffin components. For present day use these gasolines will require extensive reforming with its consequent yield loss in order to conform to market product specifications.
Various low pressure hydrocracking processes have also been described. For example, U.S. Pat. Nos. 3,867,277 and 3,923,640 disclose low pressure hydrocracking processes using various high boiling feedstocks, generally of high (20-40) API gravity. The use of such feeds, coupled with the relatively high levels of conversion in those processes leads to naphthas of low octane rating since the alkyl groups present in the feeds come through into the naphtha together with the relatively straight chain paraffins produced by the ring opening and cracking of the aromatics. These processes have therefore been unsatisfactory for the direct production of high octane gasoline.
Other low pressure hydrocracking processes producing aromatic products have been described in the past but their potential for producing high octane gasoline from low value, refractory cracking oils has not been appreciated. For example, U.S. Pat. No. 4,435,275 describes a method for producing aromatic middle distillates such as home heating oil from high gravity feeds under relatively low conversion conditions but with the objective of producing low-sulfur middle distillates, octane numbers of only about 78 (R+0) are reported.
Because highly aromatic feeds, described above, contain relatively bulky bicyclic and polycyclic components the catalyst should have a pore size which is sufficiently large to admit these materials to the interior structure of the catalyst where cracking can take place. It is noted that the amorphous hydrocracking catalysts such as alumina and silica-alumina are aromatic selective and have a large pore size, although they are not preferred. A pore size of at least about 7.4 Angstroms (corresponding to the pore size of the large pore size zeolites X and Y) is sufficient for this purpose but because the end point of the feed is limited, the proportion of bulky, polynuclear aromatics is quite low and for this reason, very large pore sizes greatly exceeding those previously mentioned are not required. Crystalline zeolite catalysts which have a relatively limited pore size range, as compared to the so-called amorphous materials such as alumina or silica-alumina, may therefore be used to advantage in view of their activity and resistance to poisoning. Catalysts having aromatic selectivity, i.e. which will crack aromatics in preference to paraffins are preferred because of the highly aromatic character of the feed.
Preferred hydrocracking catalysts, for the foregoing feeds, are the crystalline catalysts, generally the zeolites, and, in particular, the large pore size zeolites having a Constraint Index less than 2. For purposes of this invention, the term "zeolite" is meant to represent the class of porotectosilicates, i.e., porous crystalline silicates, that contain silicon and oxygen atoms as the major components. Other components are also present, including aluminum, gallium, iron, boron and the like, with aluminum being preferred in order to obtain the requisite acidity. Minor components may be present separately, in mixtures in the catalyst or intrinsically in the structure of the catalyst.
A convenient measure of the extent to which a zeolite provides control to molecules of varying sizes to its internal structure is the Constraint Index of the zeolite. Zeolites which provide a highly restricted access to and egress from its internal structure have a high value for the Constraint Index, and zeolites of this kind usually have pores of small size, e.g., less than 5 Angstroms. On the other hand, zeolites which provide relatively free access to the internal zeolite structure have a low value for the Constraint Index and usually pores of large size, e.g., greater than 8 Angstroms. The method by which Constraint Index is determined is described fully in U.S. Pat. No. 4,016,218, to which reference is made for details of the method. A Constraint Index of less than 2 and preferably less than 1 is a characteristic of the hydrocracking catalysts used in the present process.
Constraint Index (CI) values for some typical large pore materials are shown in Table 2 below:
TABLE 2 ______________________________________ Constraint Index CI (Test Temperature) ______________________________________ ZSM-4 0.5 (316.degree. C.) ZSM-20 0.5 (371.degree. C.) TEA Mordenite 0.4 (316.degree. C.) Mordenite 0.5 (316.degree. C.) REY 0.4 (316.degree. C.) Amorphous Silica-Alumina 0.6 (538.degree. C.) Dealuminized Y (Deal Y) 0.5 (510.degree. C.) Zeolite Beta 0.6-2 (316.degree.-399.degree. C.) ______________________________________
The nature of the CI parameter and the technique by which it is determined admit of the possibility that a given zeolite can be tested under somewhat different conditions and thereby exhibit different Constraint Indices. Constraint Index may vary with severity of operation (conversion) and the presence or absence of binders. Other variables, such as crystal size of the zeolite, the presence of occluded contaminants, etc., may also affect the Constraint Index. It may be possible to so select test conditions, e.g., temperatures, as to establish more than one value for the Constraint Index of a particular zeolite, as with zeolite beta. A zeolite is considered to have a Constraint Index within the specified range if it can be brought into the range under varying conditions.
The large pore zeolites, i.e., those zeolites having a Constraint Index less than 2 have a pore size sufficiently large to admit the vast majority of components normally found in the feeds. These zeolites are generally stated to have a pore size in excess of 7 Angstroms and are represented by zeolites having the structure of, e.g., Zeolite Beta, Zeolite X, Zeolite Y, Faujasite, Ultrastable Y (USY), Dealuminized Y (Deal Y), Mordenite, ZSM-3, ZSM-4, ZSM-18 and ZSM-20. Zeolite ZSM-20 resembles faujasite in certain aspects of structure, but has a notably higher silica/alumina ratio than faujasite, as do the various forms of zeolite Y, especially USY and De-AlY. Zeolite Y is the preferred catalyst, and it is preferably used in one of its more stable forms, especially USY or De-AlY.
Zeolite ZSM-4 is described in U.S. Pat. No. 3,923,639; Zeolite ZSM-20 in U.S. Pat. No. 3,972,983; Zeolite Beta in U.S. Pat. Nos. 3,308,069 and Re 28,341; Low sodium Ultrastable Y molecular sieve (USY) is described in U.S. Pat. Nos. 3,293,192 and 3,449,070; Dealuminized Y zeolite (Deal Y) may be prepared by the method found in U.S. Pat. No. 3,442,795; and Zeolite UHP-Y is described in U.S. Pat. No. 4,401,556. Reference is made to these patents for details of these zeolite catalysts.